US2024185282A1PendingUtilityA1
Method and system for estimation of optimal markdown pricing based on customer regret
Assignee: TATA CONSULTANCY SERVICES LTDPriority: Dec 5, 2022Filed: Oct 24, 2023Published: Jun 6, 2024
Est. expiryDec 5, 2042(~16.4 yrs left)· nominal 20-yr term from priority
Inventors:Suman KumarSiddhartha SarkarAvishek Kumar ShawKrishnanugrah Panaparambil UnnikrishnanVivek Balaraman
G06Q 30/0223G06Q 30/0202G06Q 30/0206G06Q 30/0283G06Q 30/0207G06Q 30/0234G06Q 30/0201G06Q 10/087G06Q 90/00G06F 17/18
52
PatentIndex Score
0
Cited by
0
References
0
Claims
Abstract
This disclosure relates generally to estimation of optimal markdown pricing based on customer regret. Pricing is a core aspect of any retail industry, and one of the pricing strategies used by retailers to attract customers is price reduction, wherein a seasonal product is sold at a markdown price after a predictable regular sale. The existing techniques for estimating an optimal markdown price are limited to factors such as market competition, and retailer formats and do not gain deeper understanding of customers' reactions nor explicitly utilizes inventory data. The disclosure estimates optimal markdown pricing in several steps including estimation of a set of behavioral model parameters comprising a customer regret parameter and a set of reservation price parameters, determining a markdown demand, and estimating the optimal markdown price using the markdown demand and inventory data.
Claims
exact text as granted — not AI-modifiedWhat is claimed is:
1 . A processor-implemented method comprising steps of:
receiving a plurality of inputs from a plurality of sources, via one or more hardware processors, wherein the plurality of inputs is associated with seasonal fashion apparels and comprises a set of transactional information and a plurality of product information associated with a plurality of products for a current regular period, a historic regular period and a historic markdown period, a fitness threshold, a markdown sale criteria, and an inventory data; identifying a first set of products from the plurality of products using the plurality of inputs based on a fitness score via the one or more hardware processors, wherein the first set of products comprises of seasonal fashion apparels sold in the historic regular period and the historic markdown period; identifying a set of similar products from the identified first set of products, via the one or more hardware processors, based on a binary mapping technique using the plurality of product information from the current regular period; estimating a set of behavioral model parameters using the identified set of similar products, via the one or more hardware processors, wherein the set of behavioral model parameters comprises a customer regret parameter and a set of reservation price parameters; updating the set of reservation price parameters to obtain a set of scaled reservation price parameters and a potential market size, via the one or more hardware processors, using the plurality of product information for the current regular period and the historic regular period based on a scaling technique; determining a markdown demand for the plurality of products of the current regular period, via the one or more hardware processors, using the set of scaled distribution parameters, the potential market size and the customer regret parameter; and estimating an optimal markdown price for the plurality of products of the current regular period, via the one or more hardware processors, using the determined markdown demand and the inventory data.
2 . The processor implemented method of claim 1 , wherein the identification of the first set of products comprises:
identifying a set of product metric for each of the plurality of products from the plurality of product information, wherein the set of product metric comprises an average price, a standard deviation and percentage sales; computing a fitness score for each of the plurality of products using the plurality of inputs and the identified set of product metric; and identifying the first set of products using the computed fitness score based on the fitness threshold and the markdown sale criteria.
3 . The processor implemented method of claim 1 , wherein the binary mapping technique comprises of assigning an attribute matching score based on a comparison between the first set of products and the plurality of product from the current regular period, wherein the attribute matching score is assigned based on a season attribute, a gender attribute, a category attribute, a sub-category attribute, an item description attribute, and a price attribute.
4 . The processor implemented method of claim 1 , wherein the estimation of the set of behavioral model parameters comprises:
identifying a regular price, a markdown price, a regular demand, a markdown demand, a no-purchase demand, a combined regular price, a combined markdown price, a combined regular demand, a combined markdown demand and a combined no-purchase demand for the set of similar products; estimating a perceived service level and an actual service level for the historic markdown period based on a minimum estimation error, wherein the minimum estimation error is heuristic technique given by identifying a value from a range of values to give a minimum error; and estimating the customer regret parameter and the set of reservation price parameters using the combined regular price, the combined markdown price, the combined regular demand, the combined markdown demand, the combined no-purchase demand, the perceived service level and the actual service level based on a constraint non-linear optimization technique, wherein the set of reservation price parameters includes a pair of Weibull distribution parameters (λ, k) and a threshold reservation price (v 1 ).
5 . The processor implemented method of claim 1 , wherein the scaling technique comprises of multiplying a scale factor with the set of reservation price parameters to obtain a set of scaled reservation price parameters and a potential market size, wherein the scale factor is a ratio between price of current regular period and the historic regular period.
6 . The processor implemented method of claim 1 , the markdown demand is determined as follows:
N
m
(
P
m
)
=
N
new
(
F
(
v
1
new
)
-
F
(
P
m
)
)
=
N
new
(
e
-
(
P
m
λ
new
)
k
-
e
-
(
v
1
new
λ
new
)
k
)
where,
N m is the markdown demand,
P m is the markdown price,
F(x)=1−e −(x/λ new ) k is the cumulative distribution function of the reservation price,
λ new and k are Weibull distribution parameters,
N new is the potential market size, and
v 1 new is the threshold reservation price.
7 . A system comprising:
a memory storing instructions; one or more communication interfaces; and one or more hardware processors coupled to the memory via the one or more communication interfaces, wherein the one or more hardware processors are configured by the instructions to:
receive a plurality of inputs from a plurality of sources, via one or more hardware processors, wherein the plurality of inputs is associated with seasonal fashion apparels and comprises a set of transactional information and a plurality of product information associated with a plurality of products for a current regular period, a historic regular period and a historic markdown period, a fitness threshold, a markdown sale criteria, and an inventory data;
identify a first set of products from the plurality of products using the plurality of inputs based on a fitness score via the one or more hardware processors, wherein the first set of products comprises of seasonal fashion apparels sold in the historic regular period and the historic markdown period;
identify a set of similar products from the identified first set of products, via the one or more hardware processors, based on a binary mapping technique using the plurality of product information from the current regular period;
estimate a set of behavioral model parameters using the set of similar products, via the one or more hardware processors, wherein the set of behavioral model parameters comprises a customer regret parameter and a set of reservation price parameters;
update the set of reservation price parameters to obtain a set of scaled reservation price parameters and a potential market size, via the one or more hardware processors, using the plurality of product information for the current regular period and the historic regular period based on a scaling technique;
determine a markdown demand for the plurality of products from the current regular period, via the one or more hardware processors, using the set of scaled distribution parameters, the potential market size, and the customer regret parameter; and
estimate an optimal markdown price for the plurality of products from the current regular period, via the one or more hardware processors, using the determined markdown demand and the inventory data.
8 . The system of claim 7 , wherein the identification of the first set of products comprises:
identifying a set of product metric for each of the plurality of products from the plurality of product information, wherein the set of product metric comprises an average price, a standard deviation and percentage sales; computing a fitness score for each of the plurality of products using the plurality of inputs and the identified set of product metric; and identifying the first set of products using the computed fitness score based on the fitness threshold and the markdown sale criteria.
9 . The system of claim 7 , wherein the binary mapping technique comprises of assigning an attribute matching score based on a comparison between the first set of products and the plurality of product from the current regular period, wherein the attribute matching score is assigned based on a season attribute, a gender attribute, a category attribute, a sub-category attribute, an item description attribute and a price attribute.
10 . The system of claim 7 , wherein the estimation of the set of behavioral model parameters comprises:
identifying a regular price, a markdown price, a regular demand, a markdown demand, a no-purchase demand, a combined regular price, a combined markdown price, a combined regular demand, a combined markdown demand and a combined no-purchase demand for the set of similar products; estimating a perceived service level and an actual service level for the historic markdown period based on a minimum estimation error, wherein the minimum estimation error is heuristic technique given by identifying a value from a range of values to give a minimum error; and estimating the customer regret parameter and the set of reservation price parameters using the combined regular price, the combined markdown price, the combined regular demand, the combined markdown demand, the combined no-purchase demand, the perceived service level and the actual service level based on a constraint non-linear optimization technique, wherein the set of reservation price parameters includes a pair of Weibull distribution parameters (λ, k) and a threshold reservation price (v 1 );
11 . The system of claim 7 , wherein the scaling technique comprises of multiplying a scale factor with the set of reservation price parameters to obtain a set of scaled reservation price parameters and a potential market size, wherein the scale factor is a ratio between price of current regular period and the historic regular period.
12 . The system of claim 7 , the markdown demand is determined as follows:
N
m
(
P
m
)
=
N
new
(
F
(
v
1
new
)
-
F
(
P
m
)
)
=
N
new
(
e
-
(
P
m
λ
new
)
k
-
e
-
(
v
1
new
λ
new
)
k
)
where,
N m is the markdown demand,
P m is the markdown price,
F(x)=1−e −(x/λ new ) k is the cumulative distribution function of the reservation price,
λ new and k are Weibull distribution parameters,
N new is the potential market size, and
v 1 new is the threshold reservation price.
13 . One or more non-transitory machine-readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors cause:
receiving a plurality of inputs from a plurality of sources, wherein the plurality of inputs is associated with seasonal fashion apparels and comprises a set of transactional information and a plurality of product information associated with a plurality of products for a current regular period, a historic regular period and a historic markdown period, a fitness threshold, a markdown sale criteria, and an inventory data; identifying a first set of products from the plurality of products using the plurality of inputs based on a fitness score, wherein the first set of products comprises of seasonal fashion apparels sold in the historic regular period and the historic markdown period; identifying a set of similar products from the identified first set of products, based on a binary mapping technique using the plurality of product information from the current regular period; estimating a set of behavioral model parameters using the identified set of similar products, wherein the set of behavioral model parameters comprises a customer regret parameter and a set of reservation price parameters; updating the set of reservation price parameters to obtain a set of scaled reservation price parameters and a potential market size, using the plurality of product information for the current regular period and the historic regular period based on a scaling technique; determining a markdown demand for the plurality of products of the current regular period, using the set of scaled distribution parameters, the potential market size and the customer regret parameter; and estimating an optimal markdown price for the plurality of products of the current regular period, using the determined markdown demand and the inventory data.
14 . The one or more non-transitory machine-readable information storage mediums of claim 13 , wherein the identification of the first set of products comprises:
identifying a set of product metric for each of the plurality of products from the plurality of product information, wherein the set of product metric comprises an average price, a standard deviation and percentage sales; computing a fitness score for each of the plurality of products using the plurality of inputs and the identified set of product metric; and identifying the first set of products using the computed fitness score based on the fitness threshold and the markdown sale criteria.
15 . The one or more non-transitory machine-readable information storage mediums of claim 13 , wherein the binary mapping technique comprises of assigning an attribute matching score based on a comparison between the first set of products and the plurality of product from the current regular period, wherein the attribute matching score is assigned based on a season attribute, a gender attribute, a category attribute, a sub-category attribute, an item description attribute, and a price attribute.
16 . The one or more non-transitory machine-readable information storage mediums of claims 13 , wherein the estimation of the set of behavioral model parameters comprises:
identifying a regular price, a markdown price, a regular demand, a markdown demand, a no-purchase demand, a combined regular price, a combined markdown price, a combined regular demand, a combined markdown demand and a combined no-purchase demand for the set of similar products; estimating a perceived service level and an actual service level for the historic markdown period based on a minimum estimation error, wherein the minimum estimation error is heuristic technique given by identifying a value from a range of values to give a minimum error; and estimating the customer regret parameter and the set of reservation price parameters using the combined regular price, the combined markdown price, the combined regular demand, the combined markdown demand, the combined no-purchase demand, the perceived service level and the actual service level based on a constraint non-linear optimization technique, wherein the set of reservation price parameters includes a pair of Weibull distribution parameters (λ, k) and a threshold reservation price (v 1 ).
17 . The one or more non-transitory machine-readable information storage mediums of claims 13 , wherein the scaling technique comprises of multiplying a scale factor with the set of reservation price parameters to obtain a set of scaled reservation price parameters and a potential market size, wherein the scale factor is a ratio between price of current regular period and the historic regular period.
18 . The one or more non-transitory machine-readable information storage mediums of claims 13 , the markdown demand is determined as follows;
N
m
(
P
m
)
=
N
new
(
F
(
v
1
new
)
-
F
(
P
m
)
)
=
N
new
(
e
-
(
P
m
λ
new
)
k
-
e
-
(
v
1
new
λ
new
)
k
)
where,
N m is the markdown demand,
P m is the markdown price,
F(x)=1−e −(x/λ new ) k is the cumulative distribution function of the reservation price,
λ new and k are Weibull distribution parameters,
N new is the potential market size, and
v 1 new is the threshold reservation price.Join the waitlist — get patent alerts
Track US2024185282A1 — get alerts on status changes and closely related new filings.
We store only your email — no account needed. See our privacy policy.